Core Entrustables in Clinical Pharmacology: Pearls for Clinical Practice
`
`Pharmacokinetics and Pharmacodynamics
`for Medical Students: A Proposed
`Course Outline
`
`The Journal of Clinical Pharmacology
`2016, 56(10) 1180–1195
`C(cid:2) 2016, The American College of
`Clinical Pharmacology
`DOI: 10.1002/jcph.732
`
`David J. Greenblatt, MD,1 and Paul N. Abourjaily, PharmD2
`
`Keywords
`pharmacokinetics, pharmacodynamics, drug interactions, medical
`
`The discipline of pharmacokinetics (PK) applies math-
`ematical models to describe and predict the time
`course of drug concentrations and drug amounts in
`body fluids.1–3 Pharmacodynamics (PD) applies sim-
`ilar models to understand the time course of drug
`actions on the body. Clinicians are most concerned with
`pharmacodynamics—they want to know how drug
`dosage, route of administration, and frequency of ad-
`ministration can be chosen to maximize the probability
`of therapeutic success while minimizing the likelihood
`of unwanted drug effects. However the path to pharma-
`codynamics comes via pharmacokinetics. Because drug
`effects are related to drug concentrations, understand-
`ing and predicting the time course of concentrations
`can be used to help optimize therapy.
`The link of drug dosage to drug effect involves
`a sequence of events (Figure 1). Even when a drug
`is administered directly into the vascular system, the
`drug diffuses to both its pharmacologic target receptor
`and to other peripheral distribution sites where it does
`not have the desired activity but may exert toxic ef-
`fects. Simultaneously, the drug undergoes clearance by
`metabolism and excretion. After oral administration,
`the situation is more complex, since the drug must
`undergo dissolution and absorption, then survive first-
`pass metabolism in the liver, before reaching the sys-
`temic circulation. Pharmacokinetics provides a ratio-
`nal mathematical framework for understanding these
`concurrent processes, and facilitates achieving optimal
`clinical pharmacodynamic effects more efficiently than
`trial and error alone.
`The following 3 clinical vignettes illustrate how
`familiarity with principles of pharmacokinetics and
`pharmacodynamics can facilitate optimal understand-
`ing and prediction of drug effects in human subjects
`and patients.
`
`Clinical Vignettes
`Case 1
`A 30-year-old man has been extensively evaluated for
`recurrent supraventricular tachycardia (SVT), which is
`associated with palpitations and dizziness. No iden-
`tifiable cardiac disease is evident, and other medical
`diseases have been excluded.
`The treating physician elects to start therapy with
`digitoxin, 0.1 mg daily. One week later the patient is seen
`again, and states that episodes of SVT are reduced in
`number. The plasma digitoxin level is 8 ng/mL (usual
`therapeutic range, 10–20 ng/mL). The dose is increased
`to 0.2 mg/day. At a follow-up visit 7 days later, the
`patient claims that symptoms attributable to SVT have
`disappeared completely. The plasma digitoxin level is
`17.4 ng/mL. The patient continues on 0.2 mg/day of
`digitoxin.
`One month later the patient sees the physician on
`an urgent basis. He has diminished appetite and waves
`
`1Program in Pharmacology and Experimental Therapeutics, Sackler
`School of Graduate Biomedical Sciences, Tufts University School of
`Medicine, Boston, MA, USA
`2Departments of Pharmacy and Medicine, Tufts Medical Center, Boston,
`MA, USA
`
`Submitted for publication 26 February 2016; accepted 3 March 2016.
`
`Corresponding Author:
`David J. Greenblatt, MD, Tufts University School of Medicine, 136
`Harrison Avenue, Boston, MA 02111
`Email: DJ.Greenblatt@Tufts.edu
`
`This proposal was prepared under the guidance of the Special Com-
`mittee for Medical School Curriculum of the American College of
`Clinical Pharmacology. The proposal
`is intended as a resource for
`the Association of American Medical Colleges as it revises its Core
`Entrustable Professional Activities for Entering Residency, Curriculum
`Developer’s Guide.
`
`1
`
`TEVA1058
`
`

`

`Greenblatt and Abourjaily
`
`1181
`
`Figure 1. Sequence of events between intravenous or oral adminis-
`tration of a drug and the drug’s interaction with the target receptor
`mediating pharmacologic action.This type of schematic diagram has been
`attributed to Dr. Leslie Z. Benet. The segment above the dashed line is
`the pharmacokinetic component—“what the body does to the drug.”
`Below the dashed line is the pharmacodynamic component—“what the
`drug does to the body” (from reference 2, with permission).
`
`of nausea. The electrocardiogram shows T-wave abnor-
`malities, and the plasma digitoxin level is 31.3 ng/mL.
`
`Discussion—Case 1
`Digitoxin is seldom used in contemporary therapeutics,
`but case 1 nonetheless illustrates 2 important principles:
`dose proportionality, and attainment of steady-state.
`The rate of attainment of the steady-state condition
`after initiation of multiple-dose treatment is dependent
`on the half-life of the particular drug. In the case of
`digitoxin, the half-life is about 7 days, implying that 3–4
`weeks of continuous treatment (without a loading dose)
`is necessary for steady-state to be reached.4 In the exam-
`ple given, the increase in dosage from 0.1 to 0.2 mg/day
`is anticipated to proportionally increase the steady-
`state concentration (Figure 2). At the daily dosage of
`0.1 mg (without a loading dose) and a half-life of 7
`days, the plasma digitoxin concentration of 8 ng/mL
`after 7 days of treatment represents 50% of the eventual
`steady-state concentration of 16 ng/mL. If the physi-
`cian had stayed with the 0.1 mg/day dosage, that steady-
`state concentration—attained after several weeks of
`treatment (4 to 5 times the half-life)—would have been
`in the therapeutic range. However the dosage was in-
`creased to 0.2 mg/day, yielding a corresponding steady-
`state concentration of 32 ng/mL, exceeding the thera-
`peutic range and producing adverse effects (Figure 2).
`
`Case 2
`A 56-year-old man has a history of grand mal seizures,
`which have been completely suppressed for the last 12
`years by phenytoin, 300 mg daily. His plasma phenytoin
`level consistently falls in the range of 12–16 μg/mL. The
`patient is able to lead a normal life, and is an excellent
`tennis player. During a particularly competitive tennis
`match, the patient injures his shoulder. That night he
`experiences severe pain, tenderness, and limitation of
`
`Figure 2. Hypothetical mean plasma concentrations of digitoxin, cor-
`responding to case 1 in the text. Digitoxin is initially given at a dosage
`of 0.1 mg per day for 1 week, after which the plasma concentration
`is 8 ng/mL (point a). The treating physician wants to increase the
`plasma concentration to a value within the usual therapeutic range
`(10–20 ng/mL). If the physician stayed with the 0.1-mg-per-day dosage,
`the eventual steady-state concentration would have been 16 ng/mL
`(dashed line)—within the desired therapeutic range. Instead, the dosage
`is increased to 0.2 mg per day on day 7. The next measured plasma
`concentration is 17.4 ng/mL 1 week later (point b), but 1 month later
`it has reached 31.3 ng/mL (point c), close to the eventual steady-state
`value of 32 ng/mL. This is well above the therapeutic range and may be
`associated with toxicity.
`
`motion. He contacts his physician. An x-ray is negative.
`The physician prescribes rest, topical heat, and aspirin,
`650 mg 4 times daily.
`At a return visit to the physician 2 days later, the
`patient is greatly improved. The physician uses that
`opportunity to do a routine check of
`the plasma
`phenytoin level, which is reported as 5 μg/mL. There
`is no evidence of recurrent seizure activity, and the
`patient insists that he is continuing to take phenytoin
`as directed (300 mg/day). The physician increases the
`dose to 500 mg/day.
`One week later the patient returns complaining of
`difficulty with balance and with fixing his eyes on
`objects. The plasma phenytoin level is 14 μg/mL.
`
`Discussion—Case 2
`In case 2, plasma protein binding of phenytoin is
`reduced by coadministration of aspirin because of
`displacement of phenytoin from plasma-binding sites
`by salicylate.5–7 This is evident as an increase in the free
`fraction (Table 1). However, the clearance of unbound
`(free) drug, and the steady-state concentration of
`unbound drug, are unchanged.8,9 As such, no change
`in clinical effect would be anticipated, and the correct
`clinical course would have been to leave the daily dosage
`at 300 mg/day. Because salicylate reduces plasma
`protein binding of phenytoin (higher free fraction in
`plasma), this has the effect of reducing total (free +
`
`2
`
`

`

`1182
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`Table 1. Total and Free (Unbound) Plasma Phenytoin Concentrations
`in Case 2
`
`Phenytoin Daily
`Dosage, and
`Cotreatment
`
`300 mg/day (no
`cotreatment)
`300 mg/day +
`aspirin
`500 mg/day +
`aspirin
`
`Total Phenytoin
`(μg/mL)
`
`Fraction Free
`
`12–16
`
`5
`
`14
`
`0.1
`
`0.3
`
`0.3
`
`Free Phenytoin
`(μg/mL)
`
`1.2–1.6
`
`1.5
`
`4.2
`
`bound) concentrations of phenytoin as well as the in-
`terpretation of these measured total concentrations.5–7
`Increasing the daily dosage of phenytoin to 500
`mg/day increases the free (unbound) concentration to
`4.2 μg/mL, which is associated with adverse effects
`despite the total concentration of 14 μg/mL.
`A second important point is the nonlinear kinetic
`profile of phenytoin.10 At daily doses in a range
`exceeding 300 mg/day, steady-state plasma concentra-
`tions increase disproportionately with an increase in
`dosage. The free phenytoin concentration is 1.5 μg/mL
`at 300 mg/day, but increases to 4.2 μg/mL with an
`increase in dosage to 500 mg/day. This property of
`phenytoin makes it difficult to titrate dosage at this
`higher dosage range.
`
`Case 3
`A fourth-year dental student is doing a clerkship in a
`dental surgeon’s practice. A healthy 34-year-old woman
`is scheduled to undergo procedures estimated to last
`approximately 2 hours. Following instillation of local
`anesthesia and prior to the start of the procedure,
`the surgeon notices that the patient still is extremely
`agitated and fearful. The surgeon administers 0.5 mg/kg
`of propofol intravenously over a 2-minute period. The
`patient becomes calm, relaxed, and falls into a light
`sleep from which she is easily roused. The surgical
`procedure is initiated and proceeds without incident for
`about 45 minutes. At this time, the patient becomes
`alert, and again is fearful and agitated. The surgeon ad-
`ministers another 0.5 mg/kg of propofol intravenously,
`the patient again becomes calm, and the surgical proce-
`dure proceeds to completion without incident.
`The student is confused. He/she asks the surgeon,
`“Why did the patient wake up after only 45 minutes?
`The half-life of propofol usually is at least 8 hours.”
`
`Discussion—Case 3
`Case 3 is an example of how the pharmacodynamic
`effects of lipophilic psychotropic drugs after single
`intravenous doses are dependent more on the rapid
`process of peripheral distribution than on clearance
`
`Figure 3. Hypothetical plasma concentrations of propofol correspond-
`ing to case 3 in the text. A 0.5 mg/kg intravenous dose of propofol is
`initially given at time zero. The plasma propofol concentration declines
`rapidly, falling below the hypothetical minimum effective concentration
`(MEC) of 200 ng/mL at 0.75 hours, at which time the patient emerges
`from the sedated condition. The rate of drug disappearance in the
`postdistributive phase would be slower (dashed line), but an additional
`0.5 mg/kg dose of propofol is required at the 0.75-hour time to maintain
`plasma concentrations above the MEC and maintain the patient in a
`sedated condition for the remainder of the procedure.
`
`or elimination. The pharmacokinetics of propofol are
`described by a 2- or 3-compartment model, in which
`the initial “distribution” phase represents rapid distri-
`bution from systemic circulation to peripheral tissues,
`followed by the terminal “elimination” phase which
`mainly reflects clearance.11–14 Although the propofol
`has not been eliminated from the body, its distribution
`from systemic circulation into peripheral tissue results
`in lower concentrations in plasma and brain, which is
`highly vascular and rapidly equilibrates with systemic
`circulation (Figure 3). As such, the patient becomes
`more alert, and a second dose is required.
`
`Components of Medical
`Education in Pharmacokinetics
`and Pharmacodynamics
`An outline of key elements of content for the teaching
`of clinical pharmacokinetics and pharmacodynamics
`to first- or second-year medical students, along with
`pertinent literature references,15–54 are presented in
`Table 2. The same or similar content can be applied to
`the teaching of graduate students at both the PhD and
`masters levels, or to postdoctoral education programs
`for house staff or practicing physicians. The material
`can be reasonably presented in 3 or 4 total lecture
`contact hours. The didactic presentations should be
`reinforced through problem sets and review sessions
`
`3
`
`

`

`Greenblatt and Abourjaily
`
`1183
`
`aimed at supporting conceptual understanding, as well
`as ensuring proficiency with pharmacokinetic calcula-
`tions and construction of graphics.
`Individual instructors can adapt the outline and
`accompanying graphics as needed, to construct specific
`lecture content consistent with their own style and
`institutional needs. An ongoing point of discussion is
`the extent to which formulas, equations, and mathemat-
`ics are needed for the teaching of pharmacokinetics.
`Student backgrounds in mathematics and their comfort
`with quantitative content vary widely. Some dislike
`and resist the mathematical content, whereas others
`welcome it. “Equation-free” pharmacokinetics is not
`realistic, but the density of equations can be managed
`such that the mathematical framework enhances con-
`
`ceptual understanding. The outline in Table 2 has that
`objective. The need for memorization of formulas and
`equations is minimal.
`Table 3 lists biomedical journals that have a focus
`on clinical pharmacokinetics and pharmacodynamics.
`Students are encouraged to consult original research
`sources when seeking information on pharmacoki-
`netic/pharmacodynamic properties or drug interactions
`involving specific drugs or drug classes. Review articles
`and secondary sources do have a role in the educational
`process, in that large amounts of data are collated and
`summarized. However, secondary sources are inevitably
`“filtered” and interpreted by their authors. Students
`need to consider the benefits and drawbacks of available
`information sources.
`
`4
`
`

`

`1184
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`Table 2. Course Outline: Pharmacokinetics and Pharmacodynamics for a Medical School Curriculum
`
`Section 1. Definition and scope
`A. Pharmacokinetics: concentration versus time
`B. Pharmacodynamics: effect versus time
`C. Kinetic-dynamic modeling: effect versus concentration
`Section 2. Value of pharmacokinetic principles in medical science15–19
`A. Choice of loading and maintenance dose
`B. Choice of frequency and route of administration
`C. Predicting the rate and extent of drug accumulation
`D. Predicting the effect of dose changes
`E. Identifying and anticipating drug interactions
`F. Identifying patient and disease factors that could alter clinical response
`G. Interpreting drug concentrations in serum or plasma19
`
`Section 3. Fundamental assumptions
`A. Proportionality cascade (Figure 4):
`–Intravascular free drug
`–Extracellular water
`–Receptor occupancy
`–Quantitative pharmacodynamic effect
`B. Concentration ranges
`–Subtherapeutic
`–Therapeutic
`–Potentially toxic
`
`Section 4. Body compartments and volumes of distribution15–17,20–22
`A. Definition of a compartment
`B. Calculation of volume of distribution derived from definition of concentration:
`Concentration = Amount
`Volume
`Volume of distribution Vd =
`
`Amount of drug in body
`Concentration in reference compartment
`C. The value and ambiguity of compartment models (hydraulic analogues)
`1. One-compartment model: Vd is unique
`2. Two-compartment model: Vd is not unique23
`D. Anatomic correlates of volume of distribution
`E. Physiochemical correlates of volume of distribution24
`
`Section 5. Exponential behavior and the meaning of half-life
`A. First-order processes:
`1. Rate is proportional to concentration
`= −kC
`dC
`dt
`2. Rate is not constant, even though k is called a “rate constant”
`B. Calculation of half-life
`= 0.693
`= In 2
`k
`k
`C. k and t1/2 are independent of route of administration
`D. Logarithmic versus linear graphs (Figure 5)
`E. Interpretation and implications of half-life
`
`t1/2
`
`Time elapsed (multiples of t1/2 )
`
`Fractional completion of process
`
`1
`2
`3
`4
`
`0.5
`0.75
`0.875
`>0.90
`
`F. Implications of first-order behavior
`–t1/2, Vd, and clearance are fixed
`–After single doses, AUC (area under the curve from time = zero to “infinity”) is proportional to dose
`–At steady state, Css (steady-state concentration) is proportional to infusion rate (or dosing rate)
`
`5
`
`

`

`1185
`
`Greenblatt and Abourjaily
`
`Table 2. Continued
`
`Section 6. Clearance of drugs and mechanisms of elimination25
`A. Model-independent definition (Figure 6):
`Clearance = Dose
`AUC
`B. Units of clearance: volume/time
`Clearance cannot exceed blood flow to clearing organ
`C. Routes of clearance
`1. Hepatic clearance: chemical modification (biotransformation)
`2. Renal clearance: excretion of intact drug
`3. All other: pulmonary, biliary, fecal, intravascular (plasma enzymes)
`D. Biologic dependence of t1/2 on both Vd and clearance
`= 0.693 x Vd
`clearance
`E. The meaning of “linear” or “dose proportional” kinetics (see also section 5F)
`
`t1/2
`
`1.
`
`k
`
`t1/2
`
`Section 7. Rapid single-dose intravenous injection
`A. One-compartment model
`= −kC
`dC
`dt
`Boundary condition: at t = 0, C= C0
`C = C0 e-kt satisfies differential equation and boundary condition
`Total AUC = C0
`k
`2. If dose = D, then
`Vd = Dose
`C0
`Note: For a 1-compartment model, Vd is unique. In practice, the best time to measure Vd is at t = 0.
`3. Monoexponential decline (Figure 5)
`= 0.693
`t1/2
`= k . Vd = k . Dose
`4. Clearance = Dose
`C0
`AUC
`5. Correct biologic relation among t1/2 , Vd, and clearance:
`= 0.693 x V d
`C lear ance
`6. Graphical approach to problem solving
`B. Two-compartment model16,17(Figure 7)
`1. Simultaneous differential equations with boundary condition:
`C = C0 at t = 0
`Solution: C = Ae-αt + Be-βt
`A, B, α, β are HYBRID
`+ B
`Total AUC = A
`β
`
`α
`
`2. Volumes of distribution20–22
`a. “Central” compartment
`V1 = Dose
`= Dose
`A + B
`C0
`b. “Total” volume of distribution
`Vd =
`= Clearance
`Dose
`β × AUC
`β
`
`t1 /2
`
`3. Half-lives
`a. Distribution half-life (cannot be derived from graph)
`α = 0.693
`α
`b. Elimination half-life (can be derived from graph)
`β = 0.693
`t1 /2
`β
`4. Clearance = dose/AUC = Vd · β
`
`6
`
`

`

`2-Compartment
`C = Ae-αt + Be-βt
`
`+ B
`β
`
`A α
`
`Dose
`β .AUC
`= Vd . β
`
`Dose
`AUC
`In 2
`β
`
`1186
`
`Table 2. Continued
`
`C. One-compartment approximation of the 2-compartment model
`Fundamental assumption:
`A and α (and the corresponding component of AUC) are ignored
`Consequences:
`B becomes C0
`β becomes k
`t1/2 β becomes t1/2
`
`Quantity
`
`Equation for concentration versus time
`
`Total AUC
`
`Total V d
`
`Clearance
`
`Elimination half-life dose
`
`D. Graphical approach to problem-solving
`E. Distribution versus elimination as the determinant of duration of action (Figures 3 and 7)
`
`Section 8. Multiple-dose kinetics: continuous intravenous infusion (without a loading dose) (Figure 8)
`A. Rate of attainment of steady state:
`Depends almost entirely on t1/2 (and k) as would occur with a single dose.
`Does not depend on infusion rate (abbreviation: Q).
`Inverted exponential function:
`C = Css (1 − e-kt). This is the same k as after a single dose.
`B. Determinants of the extent of accumulation (absolute steady-state concentration) (Figures 8 and 9)
`Css = Infusion rate
`Clearance
`C. Predictable consequences of changing infusion rate or stopping infusion (Figure 10)
`1. Rate of attainment of new steady state
`2. New steady-state concentration
`
`Section 9. Drug absorption and bioavailability (Figure 11)
`A. Rate of drug absorption
`1. Lag time (due to dissolution, gastric emptying, etc.)
`2. First-order absorption
`3. Factors influencing absorption rate
`4. Implications of absorption rate
`5. Slow-release preparations
`B. Completeness of absorption (fractional absorption or absolute systemic availability)
`1. Methods of assessment—intravenous data needed (Figure 12)
`F = AUC (P.O.)
`AUC (I.V.)
`AUC must be dose-normalized and extrapolated to infinity.
`2. Mechanisms of incomplete bioavailability25–29
`a. Incomplete absorption
`- Intrinsic properties of the chemical
`- Properties of the dosage form
`b. Efflux transport
`c. Presystemic extraction (first-pass metabolism)
`- Hepatic
`- Enteric (CYP3A)
`C. Bioequivalence and the pharmacopolitics of generic substitution30,31
`1. Competing political and economic influences
`2. Fundamental premise:
`Bioequivalence implies therapeutic equivalence (not the reverse)
`3. Methods of determining bioequivalence: statistical “equivalence” of Cmax, Tmax, and AUC
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`(cid:3)
`
`1-Compartment
`C = C0e
`−kt
`(B analogous to C0,
`(cid:2)
`β analogous to k)
`C0
`(cid:2)
`k
`
`Dose
`B
`
`analogous to
`
`= Vd . k
`
`Dose
`C0
`
`Dose
`AUC
`In 2
`k
`
`(cid:3)
`
`B β
`
`analogous to
`
`7
`
`

`

`Greenblatt and Abourjaily
`
`Table 2. Continued
`
`4. Areas of uncertainty
`a. Patients versus volunteers
`b. Limits of statistical tolerance
`c. Interpretation of anecdotes
`d. Voluntary versus forced generic substitution
`e. Sustained-release preparations
`5. Special considerations for biologic therapies (biosimilars)32,33
`
`1187
`
`Section 10. Multiple-dose kinetics: discrete doses (Figure 13)
`A. There is not a unique steady-state concentration, only a mean:
`Css = AUC over dose intervel
`length of dose intervel
`B. Rate of attainment of steady-state: identical to section 8A
`C. Extent of accumulation: if each dose (D) is 100% available and given at fixed intervals (T),
`Css =
`D/T
`Clearance
`D. Interdose fluctuation (Figure 14)
`1. Estimating the degree of fluctuation
`Cmax = maximum plasma concentration over the dosage interval
`Cmin = Minimum plasma concentration over the dosage interval
`Cmax/Cmin ratio is interdose fluctuation
`2. Css stays the same as long as D/T is unchanged, but changing both D and T influences interdose fluctuation (Figure 15)
`3. Clinical benefits of “slow-release” preparations
`E. Termination of multiple dosage: the same value of k (and t1/2 ) is applicable
`F. Loading doses (DL)
`1. Determinants of when needed; benefits and disadvantages
`2. Approximate calculations:
`Given a desired Css and assuming Vd is known, DL should cause C0 for a single dose (see section 7A) to equal the target Css during continuous infusion
`(see section 8B) or Css during multiple discrete doses (see section 10C).
`3. Complications with 2-compartment model: overshoot and undershoot
`G. Relative drug accumulation: depends on the interval between doses relative to the elimination half-life
`Section 11. Nonlinear (zero-order) kinetics10,35,37 (Figure 16)
`A. Linear (first-order) kinetics (as in section 5A):
`= −kC
`dC
`dt
`Solution: C = Coe-kt
`B. Nonlinear (zero-order) kinetics
`= −k
`dC
`dt
`Solution: C = Co − kt
`C. For most affected drugs, first-order kinetic profile transitions to zero order as the concentration increases (Figure 17)
`D. Implications of transition to zero-order kinetics
`
`Section 12. Drug interactions involving altered drug clearance2,3,36–43
`A. Epidemiology of drug interactions
`B. Mechanisms of inhibition versus induction
`C. Quantitative outcome of drug interactions (Figure 18)
`D. Clinical consequences of drug interactions: depends on the quantitative magnitude of the drug interaction and the therapeutic index of the affected drug.
`Statistical significance does not imply clinical importance.
`E. In vitro prediction of clinical drug interactions: relation of inhibitor “exposure” to inhibitory “potency”
`
`Section 13. Drug therapy in vulnerable populations: the elderly44–48
`A. Epidemiology of aging
`B. Mechanisms of altered drug response in the elderly
`1. Kinetic
`2. Dynamic
`C. Consequences of impaired clearance in the elderly.
`
`Section 14. Other vulnerable or special populations
`A. Renal insufficiency
`B. Hepatic insufficiency
`C. Obesity
`D. Children
`E. Pregnancy
`
`8
`
`

`

`1188
`
`Table 2. Continued
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`Section 15. Pharmacodynamics3,51–53
`A. Definition: time course of drug effect
`B. Approaches to measuring drug effect
`1. Fully objective (blood pressure, QT interval, serum cholesterol, etc.)
`2. Partially objective (memory tests, reaction times, pain threshold, etc.)
`3. Subjective (psychiatric endpoints)
`C. Surrogate measures of drug effect (glycated hemoglobin, T-cell count, bone mineral density, intraocular pressure, etc.)
`D. Problems with pharmacodynamic measures
`1. Effects of practice, adaptation, and time
`2. Acute and chronic tolerance
`3. Fatigue
`4. Weak connection to clinical endpoints
`E. Kinetic-dynamic modeling49–54 (Figure 19)
`1. Link between concentration and effect (linear, exponential, or sigmoid-Emax)
`2. Sources of bias
`3. Modification by effect-site entry delay
`4. Interpretation of outcome
`
`Table 3. Medical and Scientific Journals Having a Focus on Clinical
`Pharmacokinetics, Drug Metabolism, and Drug Interactions
`
`Biopharmaceutics and Drug Disposition
`British Journal of Clinical Pharmacology
`Clinical Pharmacokinetics
`Clinical Pharmacology and Therapeutics
`Clinical Pharmacology in Drug Development
`Drug Metabolism and Disposition
`European Journal of Clinical Pharmacology
`International Journal of Clinical Pharmacology
`Journal of Clinical Pharmacology
`Journal of Clinical Psychopharmacology
`Journal of Pharmaceutical Sciences
`Journal of Pharmacology and Experimental Therapeutics
`Therapeutic Drug Monitoring
`Xenobiotica
`
`Figure 4. Schematic relation between drug concentrations (green
`triangles) in circulating blood, concentrations in extracellular fluid
`(ECF) surrounding the receptor site, the extent of occupancy of
`cellular receptor sites, and subsequent pharmacologic action. If receptor
`occupancy proportionally reflects blood and ECF concentrations, then
`blood concentration may serve as a surrogate measure of drug effect.
`
`9
`
`

`

`Greenblatt and Abourjaily
`
`1189
`
`Figure 5. Plasma concentrations versus time after dosage of a drug given by rapid intravenous injection, assuming an underlying 1-compartment
`pharmacokinetic model. Each time an interval equal to the half-life elapses, the concentration falls to 50% of the value at the start of the interval.
`Left: linear concentration scale; right: logarithmic concentration scale. The same declining exponential function is applicable to both graphs (see Table
`2, section 5A–E and section 7A1–3), but the logarithmic scale transforms the exponential curve into a straight line, which can be used for graphical
`calculations. Note that a logarithmic concentration scale never goes to zero.
`
`Figure 6. Plasma concentrations of the same dose of the same drug given to different individuals. The area under the plasma concentration curve
`(AUC) is shown as the shaded region. Since clearance can be calculated as administered dose divided by AUC, the subject in the left graph has lower
`clearance (higher AUC) than the subject in the right graph (proviso: in the calculation of clearance, the “dose” represents the systemically-available
`dose, and “AUC” represents the total area under the curve from time zero to “infinity”).
`
`10
`
`

`

`1190
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`Figure 7. Left: Plasma concentrations of a drug, consistent with a 2-compartment model, after a single intravenous dose. Right: Corresponding drug
`behavior in the 2-compartment model schematic. Point I: Immediately after the dose, the entire dose is confined to the central compartment, and the
`plasma concentration is maximal (C0 = 60). Point II: During the initial distribution phase (the “alpha” phase), plasma concentrations fall rapidly and
`extensively, due mainly to drug distribution from central to peripheral compartments. Clearance (irreversible elimination) contributes minimally to this
`initial decline. If the concentration falls below the minimal effective concentration (MEC) during the distribution phase, the distribution process may
`limit the duration of clinical action. Point III: This is the point termed distribution equilibrium. The distribution process is complete, and the ratio of central
`to peripheral compartment concentrations from this time forward will remain approximately constant. Point IV: The decline in plasma concentrations
`during the elimination phase (the “beta” phase) is relatively slow, and is due mainly to clearance (irreversible elimination). (see Table 2, section 7B)
`
`Figure 8. A continuous zero-order (fixed-rate) intravenous infusion of
`a drug obeying 1-compartment kinetics is started at time zero without
`a loading dose. The concentration rises in exponential fashion until the
`steady-state condition is reached (dashed lines). The actual steady-state
`concentration (Css) at an infusion rate of Q is 3 units. If the infusion
`rate were 2 × Q, Css would be 6 units. However the rate of attainment
`of steady state is the same in both cases, being dependent only on the
`half-life (from reference 2, with permission).
`
`Figure 9. Relation between zero-order infusion rate (Q, X axis) and
`steady-state plasma concentration (Css, Y axis) for a drug having linear
`(first-order) kinetic properties, as shown in Figure 8. Css has a direct
`linear relation to Q (see Table 2, sections 5F and 8B). The slope of the
`line is 1/clearance.
`
`11
`
`

`

`Greenblatt and Abourjaily
`
`1191
`
`Figure 10. Left: A drug is given by continuous zero-order intravenous infusion (fixed rate of Q) starting at time zero. The plasma concentration
`ascends to Css (4 units) in exponential fashion, with attainment of steady state more than 90% complete after 4 × t1/2. This is similar to Figure 8.
`Right: After steady state is reached, the infusion rate is changed (arrow), and the time scale “resets” to zero. The rate is either increased to 2 × Q,
`decreased to 0.5 × Q, or stopped altogether. The new Css correspondingly changes to 8 units, 2 units, or 0, respectively. However, another 4 or more
`multiples of t1/2 are needed for the new steady state to be reached.
`
`Figure 11. Plasma concentrations of a drug after oral dosage at
`time zero. After a lag time (Tlag), plasma concentrations begin to rise.
`During this “absorption” phase, rates of drug entry into blood due to
`absorption exceed rates of elimination due to clearance. The maximum
`concentration (Cmax = 7 units) is reached at time Tmax (1 hour after
`dosage); at this point, instantaneous absorption and elimination rates are
`equal. Concentrations then fall, indicating that elimination rates exceed
`absorption rates. The area under the plasma concentration curve (AUC)
`is used as a surrogate for the extent of absorption (from reference 2,with
`permission).
`
`Figure 12. Mean plasma concentrations of midazolam after single
`intravenous and oral doses administered to healthy volunteers29 (*con-
`centrations were normalized to a 2-mg dose). Based on the relationship
`in section 9B1, in Table 2, the absolute bioavailability of oral midazolam
`(F) was calculated as 0.29,indicating that only 29% of an oral dose actually
`reaches the systemic circulation (from reference 2, with permission).
`
`12
`
`

`

`1192
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`Figure 13. A drug is administered as discrete oral doses at intervals equal to the half-life. Six consecutive doses were given, after which the drug
`was discontinued. The dashed line is the hypothetical curve if the same dosing rate were administered by continuous zero-order intravenous infusion.
`Note that the rate of attainment of steady state is the same in both cases.
`
`Figure 14. A dosage interval at steady state. The elimination half-life
`is assumed to be 24 hours, and the dosage interval is also 24 hours
`(as in Figure 13). The plasma concentration starts at Cmin, increases to
`Cmax, then falls again to Cmin. The mean steady-state concentration (Css,
`dashed line) is the area under the curve for 1 dose interval divided by
`the length of the interval. The interdose fluctuation is calculated as the
`Cmax/Cmin ratio. (see Table 2, section 10A–D).
`
`Figure 15. Serum concentrations of a drug at steady state if the drug
`is given using 3 different dosing schedules: either 500 mg every 24 hours,
`250 mg every 12 hours, or 125 mg every 6 hours. In each case the mean
`steady-state concentration (Css) is the same, since the [(dose)/(dose
`interval)] ratio does not change. However, the interdose fluctuation
`becomes smaller as the dose interval becomes smaller. With the once-
`daily dosing schedule, the interdose fluctuation is large, and the plasma
`concentration falls outside the therapeutic range just after and just
`before each